Intersecting roles CMS and FDA implications for pharmaceutical and device industries

Transcription

1 Intersecting roles CMS and FDA implications for pharmaceutical and device industries Peter B. Bach, MD, MAPP Senior Adviser, Office of the Administrator Centers for Medicare & Medicaid Services

2 Traditional & intersecting roles CMS Traditional: pay for healthcare services Reality: Have to make decisions about what to pay for based on evidence Collects data on healthcare events by virtue of claims Has ability of spurring creation of new knowledge, through coverage FDA Traditional: ensure safety of medical items prior to market Reality: Sets evidentiary standards Ability to mandate post-market evidentiary development Great interest in post-market surveillance for adverse events

3 Examples of intersecting activities at CMS Coverage determinations Greater clarity in recent years Not redundant with FDA decisions But not unrelated Increasing work on coverage in the context of evidence development (CED) Drawing more attention to representativeness issues in clinical trials Surveillance Claims data for health outcomes assessment (not just cost calculations) Wennberg studies variations in costs without variations in outcomes Patterns of care studies underutilization of worthwhile therapy Observaitonal data analyses can estimate treatment effects Moving forward to create or motivate surveillance systems Registries for particular conditions Claims data, including drug claims for part D

4 Coverage static decisions Evidentiary standards increasingly codified Individual coverage determinations articulate reasoning Guidance documents articulate principles Standards not the same as FDA CMS sometimes non-covers drugs/devices despite labeling But, CMS often covers non-labeled uses Mandate: pay for items or services that are reasonable and necessary for management Evidence hierarchy topped by RCT s

5 Coverage dynamic decisions Coverage under evidence development Service/Item covered in the context of further evidence gathering Additional patient information gathered to assure appropriate use of item or service Can drive longitudinal data collection Specific investigation into impact of item or service on health being conducted as part of coverage (coverage only in a clinical trial)

6 Example of CED ICD (implantable cardiac defibrillator registry) Registry required as part of coverage Assures appropriate patients get service Longitudinal data on firing and mortality developed outside CMS Should produce valuable evidence about effectiveness and complications.

7 Another example of CED PET (positron emission tomography) scanning for cancer patients Coverage supported for all eligible patients Additional clinical data on patients required for coverage Collected through registry that is also addressing relevant clinical questions How does PET scan alter clinical mgmt

8 Another example of CED Coverage of oxygen therapy in NHLBI trial Will definitively answer whether it benefits patients with obstructive lung disease Can be used for future National Coverage Determination

9 Reasons CED gets triggered Reasonable questions need to be answered in order to make coverage determinations Patients studied not representative of Medicare population Further baseline data needed to assure appropriateness All move CMS coverage process from static to one of evidence development and surveillance

10 Surveillance Through paying based on fee-for-service, CMS gathers longitudinal health records for every beneficiary Decades of research evaluating these claims histories have demonstrated that patterns of claims can be analyzed to determine Patterns of care Impact of services

11 2 major surveillance initiatives CED through registries to establish monitoring systems (as for ICD s), prospective inception data developed with important clinical details beyond that available in claims Can overlap with FDA mandated post-market activities Use of claims for drugs and services to monitor for adverse events Fair amount of work looking at adverse events after major surgeries Some work looking at adverse events in relation to physician administered drugs (part B drugs) Next frontier is the Part D prescription drug system

12 Part D claims Surveillance systems should be large and comprehensive 1M p-years accumulated in FFS Medicare every 9 days 3M prescriptions filled daily Claims should capture impt events Medicare pays for care of most serious medical events and complications

13 Opportunities Evaluate individuals on various Part D drugs for frequency of healthcare events occurring in claims Adverse event surveillance Comparative studies between agents Assessment for rare but known side effects Evaluate adequacy of medications obtained by beneficiaries linked to diagnoses E.g. look for diabetics on and off ACEI s Evaluate impact of therapies in rarely studied populations Minorities, elderly

14 Cautions Adverse event monitoring will have to be conducted by analyzing millions of people on many millions of drugs experiencing many millions events Therefore, many many analyses will be performed any many different parts and combinations of data Epidemiologically, a risk that multiple analyses lead to false positive results How to handle this? Put systems in place to protect against false positives, like validation datasets

15 Conclusions CMS is doing a fair amount that is related to FDA s regulatory activity Clarifying evidentiary requirements drives study design during development phase Further enabling evidence development through CED enhances evidence base for particular uses and creates post-market surveillance resources Linking claims for drugs (part b and d) to other Medicare claims creates an enormous resource for adverse event monitoring.